Artificial neural network in cosmic landscape

A bstractIn this paper we propose that artificial neural network, the basis of machine learning, is useful to generate the inflationary landscape from a cosmological point of view. Traditional numerical simulations of a global cosmic landscape typically need an exponential complexity when the number of fields is large. However, a basic application of artificial neural network could solve the problem based on the universal approximation theorem of the multilayer perceptron. A toy model in inflation with multiple light fields is investigated numerically as an example of such an application.

[1]  Max Tegmark,et al.  What does inflation really predict? , 2004, astro-ph/0410281.

[2]  D. Langlois,et al.  Beauty is distractive: particle production during multifield inflation , 2011, 1106.1891.

[3]  Richard Easther,et al.  Cosmology from random multifield potentials , 2005, hep-th/0512050.

[4]  P. Hayden,et al.  Quantum computation vs. firewalls , 2013, 1301.4504.

[5]  M.H. Hassoun,et al.  Fundamentals of Artificial Neural Networks , 1996, Proceedings of the IEEE.

[6]  Roger G. Melko,et al.  Machine learning phases of matter , 2016, Nature Physics.

[7]  Sahal Yacoob,et al.  A neural network clustering algorithm for the ATLAS silicon pixel detector , 2014 .

[8]  C. Vafa.  The String Landscape and the Swampland , 2005, hep-th/0509212.

[9]  Thermal properties of an inflationary universe. , 1996, Physical review. D, Particles and fields.

[10]  T. Battefeld,et al.  Local random potentials of high differentiability to model the Landscape , 2014, 1409.5135.

[11]  Andreas Albrecht,et al.  Cosmology for Grand Unified Theories with Radiatively Induced Symmetry Breaking , 1982 .

[12]  A. Liddle,et al.  Multifield consequences for D-brane inflation , 2012, 1203.3792.

[13]  Levent Sagun,et al.  Energy landscapes for machine learning. , 2017, Physical chemistry chemical physics : PCCP.

[14]  Trans-Planckian problem of inflationary cosmology , 2000, hep-th/0005209.

[15]  Angela Spalsbury,et al.  The Joys of Haar Measure , 2014 .

[16]  A meandering inflaton , 2009, 0910.0849.

[17]  Michael R. Douglas,et al.  Computational complexity of the landscape I , 2006, ArXiv.

[18]  Fabian Ruehle.  Evolving neural networks with genetic algorithms to study the string landscape , 2017, 1706.07024.

[19]  S. Tye,et al.  THE COSMOLOGICAL CONSTANT PROBLEM AND INFLATION IN THE STRING LANDSCAPE , 2008, 0803.0663.

[20]  R. Bousso,et al.  Fast optimization algorithms and the cosmological constant , 2017, 1706.08503.

[21]  A. Haar.  Der Massbegriff in der Theorie der Kontinuierlichen Gruppen , 1933 .

[22]  Yichen Huang,et al.  Neural Network Representation of Tensor Network and Chiral States. , 2017, Physical review letters.

[23]  Yi Wang,et al.  Quasi-Single Field Inflation and Non-Gaussianities , 2009, 0911.3380.

[24]  A. Szalay,et al.  The statistics of peaks of Gaussian random fields , 1986 .

[25]  David H. Lyth,et al.  What Would We Learn by Detecting a Gravitational Wave Signal in the Cosmic Microwave Background Anisotropy , 1997 .

[26]  Dmitri Krioukov,et al.  Machine learning in the string landscape , 2017, Journal of High Energy Physics.

[27]  L. Motl,et al.  The String landscape, black holes and gravity as the weakest force , 2006, hep-th/0601001.

[28]  D. Green.  Disorder in the early universe , 2014, 1409.6698.

[29]  W. Pitts,et al.  A Logical Calculus of the Ideas Immanent in Nervous Activity (1943) , 2021, Ideas That Created the Future.

[30]  M. Douglas,et al.  Basic results in vacuum statistics , 2004, hep-th/0409207.

[31]  Yang-Hui He,et al.  Deep-Learning the Landscape , 2017, 1706.02714.

[32]  Yann LeCun,et al.  Explorations on high dimensional landscapes , 2014, ICLR.

[33]  A. Guth,et al.  Counting Vacua in Random Landscapes , 2016, 1612.05224.

[34]  J. Billingsley.  Mathematics for Control , 2005 .

[35]  George Cybenko,et al.  Approximation by superpositions of a sigmoidal function , 1992, Math. Control. Signals Syst..

[36]  F. Denef,et al.  Computational complexity of the landscape II - Cosmological considerations , 2017, 1706.06430.

[37]  P. Chaudhari,et al.  The Effect of Gradient Noise on the Energy Landscape of Deep Networks , 2015 .

[38]  R. Brandenberger,et al.  Multi-stream inflation in a landscape , 2011, 1201.0029.

[39]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[40]  G. Ellis,et al.  Universe or Multiverse , 2009 .

[41]  Kurt Hornik,et al.  Approximation capabilities of multilayer feedforward networks , 1991, Neural Networks.

[42]  W. Browder,et al.  Annals of Mathematics , 1889 .

[43]  Gary William Flake,et al.  Nonmonotonic activation functions in multilayer perceptrons , 1993 .

[44]  Siyi Zhou,et al.  Nonuniqueness of classical inflationary trajectories on a high-dimensional landscape , 2015, 1501.06785.

[45]  Liam McAllister,et al.  Charting an Inflationary Landscape with Random Matrix Theory , 2013, 1307.3559.

[46]  Peter Norvig,et al.  Artificial Intelligence: A Modern Approach , 1995 .

[47]  Shamit Kachru,et al.  De Sitter vacua in string theory , 2003, hep-th/0301240.

[48]  Vacuum configurations for superstrings , 1985 .

[49]  C. A. Wuensche,et al.  A neural-network based estimator to search for primordial non-Gaussianity in Planck CMB maps , 2014, 1409.3876.

[50]  Andrei Linde,et al.  A new inflationary universe scenario: A possible solution of the horizon , 1982 .

[51]  Thomas C. Bachlechner.  On Gaussian random supergravity , 2014, 1401.6187.

[52]  A. Starobinsky.  STOCHASTIC DE SITTER (INFLATIONARY) STAGE IN THE EARLY UNIVERSE , 1986 .

[53]  A. Guth.  Inflationary universe: A possible solution to the horizon and flatness problems , 1981 .

[54]  D. Krefl,et al.  Machine Learning of Calabi-Yau Volumes : arXiv , 2017, 1706.03346.

[55]  A. Liddle,et al.  Multi-field inflation with random potentials: field dimension, feature scale and non-Gaussianity , 2011, 1111.6646.

[56]  Understanding the landscape , 2006, hep-th/0602266.